The Campaign to Pasteurize Breast Milk

By Sylvia Onusic, PhD, CNS, LDN

According to the Ross Mothers Survey, which has been tracking breastfeeding data for years, breastfeeding at birth has increased from approximately 25 percent in the early 1970s to 75 percent in 2007. The overall benefit of feeding babies breast milk instead of formula is well established. In 2007, the U.S. Department of Health and Human Services issued a report showing that babies who are formula-fed instead of breastfed are at increased risk for asthma, acute ear infections, diarrhea, and SIDS (sudden infant death syndrome).1

But less than half of all mothers are able to breast feed their babies to age six months as recommended by the American Association of Pediatrics, and just a fifth of those breastfeed for an entire year.1

Before baby formula, “wet nurses” fed babies who could not be nursed by their mothers. Such informal breast milk sharing has gone on for thousands of years. But in these modern times when many of us don’t even know our neighbor, the Internet provides the social gathering place for meeting the traditional needs of modern women.

The four main Internet sites that provide breast milk, “Human Milk 4 Human Babies,” “Eats on Feets,” “Only the Breast,” and “Milk Share.” Emma Kwasnik, Founder of Human Milk 4 Human Babies, says that at her organization, “women graciously share their breast milk with babies and children in need in their local communities.” The visibility and popularity of these sites prompted a Columbus, Ohio, research group to examine the quality of Internet breast milk donations. Their results, which were published at the end of October 2013 in the journal Pediatrics, quickly went viral on the Internet. Lead author Sarah Keim and her team reported that the human milk samples were “highly contaminated” with pathogenic bacteria and “posed a risk for negative outcomes” for infants, especially premature or low birthweight infants.2

A Flawed Protocol?

The research group solicited samples of breast milk from the four main websites providing this service and ended up with one hundred one samples of milk. The researchers instructed the sellers to send the samples to an anonymous rented post office box. They then compared the quality of these samples to twenty unpasteurized samples of unknown age from the local milk bank, which served as a control group.

The researchers “analyzed all the samples in a single batch.” We don’t know whether this refers to both the milk bank samples and the Internet samples, or just the Internet samples, because the text does not specify. However, Table 1 in the paper shows separate results for the Internet samples and the milk bank samples.

When testing for HIV and cytomegalovirus, the laboratory technicians used the latest in molecular biology technology, DNA testing; for the bacteria, they used the traditional method of growing bacteria on laboratory plates. They then calculated the prevalence of the bacteria which they grew on the plates, and used statistical analysis to compare the mean and prevalence of the bacteria grown on the specific plates with those from the milk bank samples.

According to their analysis for prevalence, gram-negative bacteria and Staphylococcus spp were the only bacteria that were significantly more prevalent in the Internet breast milk compared with those of the milk bank. The average number of selected bacteria was significantly higher in the Internet samples compared to the milk bank samples, except for the Streptococcus species.2

Media Portrayal

The New York Times, USA Today, NBC News, even PBS, and many bloggers served up the same version of the story. The abstract of the study was readily available on the Internet but the article was available only by purchase. It seemed that many of the media reporters and bloggers did not bother to read the entire text. With the headline “Breast Milk Donated or Sold Online Is Often Tainted, Study Says,” The New York Times referred to the sale and donation of human breast milk on the Internet, as a “cottage industry.” Their story relied almost exclusively on an interview with the lead researcher, Sarah Keim.3

According to the study abstract, “most (74 percent) Internet milk samples were colonized with Gram-negative bacteria . . . and exhibited higher mean total aerobic, total Gram-negative, coliform, and Staphylococcus sp (species) counts than milk bank samples. No samples were HIV type 1 RNA-positive; 21 percent of Internet samples were cytomegalovirus DNA-positive.”2

According to Keim, the researchers found that 64 percent of the samples from the Internet sellers were “contaminated” with staph, 36 percent with strep, and almost three-quarters with other bacterial species including salmonella. Keim told The New York Times that “most staph and strep are harmless at normal levels,” but that salmonella “doesn’t belong in milk at all.”3

Alive With Probiotic Bacteria

What’s misleading about this study is the fact that the researchers don’t tell us the strain of the bacteria, just the general species that was found in the milk, such as Streptococcus sp. (The designation for species normally used is spp not sp as indicated in the paper.) The vast majority of these general strains are benign.

In fact, breast milk is alive with probiotic bacteria. A 2012 study found a large microbial diversity in colostrum and breast milk. Around seven hundred species populate the breast milk microbiome. The most common bacterial genera in the colostrum samples in this study were weissella, leuconostoc, staphylococcus, streptococcus and lactococcus.4 Weissella is a lactic acid bacteria related to lactobacillus. The most common bacteria found in breast milk, probiotic bacteria, as well as commensals, were, in the Staphylococcus spp. (staph) species (spp): S. epidermidis, S. hominis, S. capitis and S. aureus; and in the Streptococcus (strep) spp: S. salivarius, S. mitis, S. parasanguis, and S. peores.5

The Wall Street Journal reported that, “Some infection-causing bacteria, including E. coli, staphylococcus and streptococcus, were found in 72 percent of the samples.” How they reached this conclusion is unknown because the study itself did not actually report the type of staph and strep in the milk, which usually contains several staph and strep species as defined above.6

Bloggers expressed alarm that Escherichia coli (E.coli) was found in the breast milk. E. coli is a gram-negative bacteria commonly found in the intestine.7 As part of the normal gut flora it produces vitamin K2 and controls pathogenic bacteria within the intestine.8 Keim confirmed in her interview with The New York Times that E. coli is “normal” (she did not say “benign”) and warned that the bacteria “probably came from improper washing of hands, breast pumps or milk containers and then proliferated during storage and shipping.”3

Regarding the cytomegalovirus found in the Internet samples, the CDC (Centers for Disease Control) says that “cytomegalovirus (CMV) can be transmitted to infants via contact with maternal genital secretions during delivery or through breast milk,” and that these infections don’t harm the infant. Healthy babies who contract CMV after birth “generally have few, if any, symptoms or complications from the infection.”9


In response to the Keim article, Joseph A. Ladapo, assistant professor of medicine at the New York University School of Medicine points out the main weakness of the study: it compared the highly selective one hundred one Internet milk samples to only twenty milk bank samples that served as the control group, a tactic that greatly reduces the statistical power, that is, the ability to make correct predictions. He points out that a larger number of Internet milk samples and equal number of milk bank samples might have yielded much different results. The number of control samples should equal the samples to be tested. Ladapo says that, “Some parents (including this author) go to considerable lengths to provide their infants with human breast milk because of the body of evidence supporting its health benefits.”

He suggests a study using real parents and their babies and other positive studies to assist parents in protecting and supporting their children. He also throws out a very provocative statement regarding contamination of the samples occurring in the medical setting: “Studies in medical settings support the notion that some hygienic practices are often difficult to sustain”10 or, in plain English, were the samples contaminated at the lab?

Regarding three samples that contained salmonella, “not normally found in breast milk,”3 the question of contamination in the hospital rears its ugly head. With freezing, thawing and refreezing, and later processing, some contamination certainly could have occurred.

Indeed, the way that the samples were procured and handled certainly poses some serious questions. The “Internet samples” were “stored at -20 degrees Centigrade until analyzed within two months.”2 The article does not tell us the condition of the samples when they arrived, considering they may have lingered in the rental mail box for an undefined period of time until pickup. “How diligent were the staff in picking up those samples every day right after they arrived? We would assume that the samples were frozen when sent because moms normally freeze excess breast milk.”11 Researchers left it up to the sellers to choose the packing materials, also jeopardizing the results of the study. We don’t know whether frozen samples remained frozen, were allowed to thaw, or were at room temperature. One thing is certain, after arrival the samples were returned to the freezer for up to two months.2 Freezing and thawing and refreezing would alter bacterial content.

One blogger who criticized the study had apparently read the paper. Sarah, mother of three breastfed children and author of the blog Nurshable: Joy in Gentle Parenting (established in 2011 with twenty to thirty thousand viewers per month), says that “re-freezing the milk goes against every breast milk handling guideline out there.” She also warns that the methods used by the researchers to obtain the milk samples would raise a red flag for all those who donate breast milk. “The researchers refused to speak in person or over the phone, dodged questions about the baby, requested only a small amount of milk, did not designate shipping and storage conditions.” One of the most important steps, a typical conversation about the recipient baby, did not take place. Sarah feels that in this study real life conditions were not met, that the researchers purposely excluded responsible donors, and that the study did much damage to the breast milksharing community.11

The authors of the paper admit that their procedures to maintain anonymity may have biased the results because women who asked about a recipient infant or asked for telephone communication were excluded from the study. They didn’t collect useful information from the sellers, like whether the milk was ever frozen, or even information about the seller’s health. They also admitted that they didn’t know the age of the twenty milk bank samples used as controls, and that some could have been shipped just like the Internet samples were, which would mean that they could have been frozen, thawed and refrozen as well.2

The Push To Pasteurize

In the fall of 2010, the FDA issued a warning to moms to avoid feeding babies another woman’s breast milk because of possible body fluid contamination.12 But to date there are no reported cases of infection from breast milk acquired online.13 And no examples of babies who have become ill consuming Internet breast milk were given in the study.2

Emma Kwasnica, founder of the Human Milk 4 Human Babies Global Network, said that in her experience, sellers and buyers meet and “often form close relationships.” “With community milk-sharing, you know exactly who your milk is going to,” she said. The Human Milk 4 Human Babies network now includes one hundred thirty Facebook communities in fifty-two countries. “If babies were getting sick from this milk, we would know,” she said.6

But study author Sarah Keim appeals to fear: “Infants consuming this milk are at risk for negative outcomes, particularly if born preterm or are medically compromised,“ she warns. “The potential risk of milk sharing to infant health needs to be further examined related to other risks,” such as toxins, pharmaceutical and drug exposure. Then comes the push to pasteurize. “Milk banks usually pasteurize milk, which is largely effective in limiting the risk of bacterial and viral illness.” She says that human milk sharing via the Internet has the potential “to cause infectious disease” and, in these samples, the bacterial contamination points to “inadequate hygiene at the point of milk expression. . . .”2

The study conclusions don’t provide many options for those moms who need to find sources of breast milk.

Should Breast Milk Be Pasteurized?

What Keim doesn’t tell us, and which she should know, is that studies show that raw breast milk is both safer and more effective than pasteurized milk. In a randomized, controlled trial of two hundred twenty-six babies, infections were significantly higher in babies given pasteurized human milk and formula (33 percent), pasteurized human milk (14 percent) and raw human milk plus formula (16 percent) than the babies given raw human milk (10 percent).14 This point needs to be emphasized. Babies given raw human milk had the lowest rate of infection; babies given pasteurized human milk plus formula (which is the standard of care) had the highest rate of infection.

In another study, very low birth weight infants had significantly more rapid weight gain when fed their own mother’s untreated milk compared to babies fed pooled pasteurized mother’s milk. The authors speculated that the poorer results from pasteurization were due to the fact that heat treatment destroys heat-labile enzyme lipase.15

Mother’s milk should be frozen, not pasteurized. But almost all milk banks pasteurize to 144.5 degrees Fahrenheit. Hospitals then charge four to six dollars per ounce for breast milk. In order to sell the milk to hospitals, the milk must have a certain protein, fat, lactose and calorie profile calculated by the “milkoscan FT 120.”13 The milk is available only by prescription and mostly for premature infants.

Lori Feldman-Winter, a professor of pediatrics at Cooper University Hospital in Camden, N.J., who was quoted on many web stories related to the study, says that “greater support from doctors, hospitals and employers would help more mothers breast-feed their own babies.”6

That topic is a story in itself. One thing is certain, all health professionals must be involved in actively promoting a mother’s right to nurse her baby and providing information on appropriate healthy alternatives, whether raw human breast milk or homemade formula based on raw milk.  Let’s remember why breastfeeding is so important: because breast milk is raw milk, teaming with rich biodiversity.


1. The Surgeon General’s Call to Action to Support Breastfeeding. 2011. Public Health Service. U.S. Department of Health and Human Services. Accessed 11-23-2013.

2. Keim SA et al. Microbial Contamination of Human Milk Purchased Via the Internet. Pediatrics 2013; 132:5 e1227-e1235; published ahead of print October 21, 2013, doi:10.1542/peds.2013-1687.

3. Bakalar, M. Breast Milk Donated or Sold Online Is Often Tainted, Study Says. The New York Times. October 21, 2013.

4. Cabrera-Rubio R et al. The human milk microbiome changes over lactation and is shaped by maternal weight and mode of delivery. AmJClinNutr. 2012; 96 (3): 544 DOI: 10.3945/ajcn.112.037382

5. Lara-Villoslada F, Olivares M, Sierra S, Rodriguez JM, Boaz J, Xaus J. Beneficial effects of probiotic bacteria isolated from breast milk. Br J Nutr. 2007; Oct;98 Suppl 1:S96-100.

6. Beck. M. Donor Breast Milk Shown to Be Risky. Study Finds High Levels of Bacteria in Majority of Samples Obtained Online. Wall Street Journal. 10-21-2013

7. Escherichia coli”. CDC National Centers for Disease Control and Prevention. Retrieved 2013-11-25.

8. Bentley R, Meganathan R. Biosynthesis of vitamin K (menaquinone) in bacteria. Microbiol Rev. 1982. 46 (3): 241–80. 9

9. Centers for Disease Control. Cytomegalovirus (CMV) and Congenital CMV Infection, accessed November 26, 2013.

10. Ladapo, JA. Replies to Microbial Contamination of Human Milk Purchased Via the Internet. What Next for Human Breast Milk Sharing Research. Pediatrics. 2013. 132:5 11.4.2013.

11. Nurshable. Joy in Gentle Parenting. How to Not Conduct a Study about Peer Breastmilk Sharing. Accessed 10-22-13.

12. U.S. Food and Drug Administration FDA. Science and Research. Use of Donor Milk.  Accessed 11-12-2013

13. Dutton J. Liquid Gold: The Booming Market for Human Breast Milk. May 17, 2011.

14. Narayanan I et al. Randomised controlled trial of effect of raw and holder pasteurised human milk and of formula supplements on incidence of neonatal infection. Lancet. 1984 Nov 17;2(8412):1111-3.

15. Stein H et al. Pooled pasteurized breast milk and untreated own mother’s milk in the feeding of very low birth weight babies: a randomized controlled trial. J Pediatr Gastroenterol Nutr. 1986 Mar-Apr;5(2):242-7.


The Industrial Replacement for Breast Milk

Prolacta Bioscience in Monrovia, California is the first company to develop and sell human milk product for a profit. Prolact +4 H2MF is an enhanced breast milk product in the form of a “syrupy fortifier,” concentrated to a 10-1 ratio to be used in the hospital, costing almost two hundred dollars per ounce for premature and low weight babies. A ten-week supply of the syrup costs over ten thousand dollars per baby and is typically covered by insurance, while breast milk supplied by a milk bank is not. Other versions are available in more “souped up” super strength, +6, +8, and +10. The product is used to fortify mom’s breast milk.

The company started in 1999 and now has a client roster of over fifty hospitals. Prolacta chief executive, Scott Elster, says that “sales grew by 200 percent in 2010.” He is confident that “sales will continue to grow at this pace.” It remains to be seen how much babies will benefit from this new industrial supplementary product with a shelf life of three years at room temperature.2

The company accepts donor milk but does not financially compensate the donors because the final product may be somehow tainted. No matter, it seems to have adequate supplies of breast milk. They do pay a team of aggressive recruiters who travel the county searching out donating moms for Prolacta’s “Milkin’ Mamas” program through contact at hospitals, baby supply stores, pediatrician’s offices and women’s events.

Elster is confident that his supply of milking moms will continue as volunteers—he says that, “our main hold-up is educating the market”—that is, persuading more hospitals to buy Prolacta’s superserum, made by separating the milk, much like in the industrial milk process for cow’s milk, using the protein part and then adding back some fat.1 His other challenge is persuading more mothers to donate.

The Prolacta website (, lists the studies the company has financed in cooperation with hospitals and universities to validate their product. Like pharmaceutical companies, their chief researchers also speak at national conferences, such as the American Academy of Pediatrics.

After the research article, “Microbial Contamination of Human Milk Purchased Via the Internet,” from the Columbus, Ohio team went viral on the internet,, one of the four major internet sites that provide breast milk, decided to pull out of the game, discontinue its services and to form a partnership with Medolac Laboratories, based in Lake Oswego, Oregon, another of the for-profit breast-milk companies. The Mothers Milk Cooperative is another of their clients.2 It seems that will funnel its breastfeeding moms to Medolac, which plans to screen donors and buy their milk for one dollar per ounce.

Lactating moms, with the best intentions, thinking that their milk is going to help premature babies, continue to donate. But their sacred milk is used to produce an almost obscene product that has been patented, pasteurized and highly refined, for the benefit of stockholders.


1. Dutton J. Liquid Gold: The Booming Market for Human Breast Milk, May 17, 2011.

2. Beck. M. Donor Breast Milk Shown to Be Risky. Study Finds High Levels of Bacteria in Majority of Samples Obtained Online. Wall Street Journal. 10-21-2013


About the Author

Sylvia OnusicSylvia Onusic, PhD, CNS, LDN, a licensed nutritionist, writer and researcher, spoke at Wise Traditions 2013 on “Traditional Foodways of Slovenia.” She is an active contributor to the Weston A. Price Foundation journal Wise Traditions and She holds a BS in home economics, foods and nutrition education, an MS in the field of health administration and policy, and a PhD in public health education. She completed the certified nutrition specialist credential in December 2012. Sylvia is a board certified nutrition specialist (CNS) and licensed dietitian nutritionist (LDN), as well as a member of the American College of Nutrition. While a Fulbright Scholar in the Republic of Slovenia in the field of public health, she completed research at the National Institute of Public Health, and later was employed at the Ministry of Health for six years. She can be reached at


2 thoughts on “The Campaign to Pasteurize Breast Milk

  1. Dear Ms. Kwasnica,
    I am delighted with Your article as well as the interest in the expressed human milk. There is a fact that the potential health effect of various type of bacteria in expressed human milk is unclear. Many of microbiological criteria in human milk banks are derived from standards used in the dairy plants. However, how to change the public health officer’s opinion that expressed human milk is similar to cow’s milk in dairy? This leads to pointless checks by the inadequate standards! How to do it in the countries where the tradition of different regulations is strong?
    Thanks for Your reply.
    With regards.
    Ales Perina

  2. As a breastfeeding mom to 3 babies, this just ticks me off. Pasturization for bmilk is rediculous and I would be super pissed off if I had spent time pumping milk for a baby to be thrown away for a science experiment.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.